INDEX

Explanations

references to educational assignments and related coursework

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

expandindo

-0.74

 AssemblyCulture

-0.63

Hochspringen

-0.52

Legături

-0.50

BeginContext

-0.48

 kaynağından

-0.47

 AssemblyVersion

-0.47

IntoConstraints

-0.45

UpInside

-0.45

 découvrir

-0.44

POSITIVE LOGITS

 assignments

0.96

 Assignment

0.93

 assignment

0.93

 homework

0.91

 Assignments

0.90

 Homework

0.89

homework

0.86

Assignment

0.86

 assig

0.82

Homework

0.82

Activations Density 0.329%

references to educational assignments and related coursework

No Comments

No Known Activations